Flame Depth Research Articles

Investigation on the restrain effect of air curtain to the spilled flame and its heat flux of a compartment-facade fire

This paper investigates the restrain effect of air curtain to the spilled flame and its heat flux of a compartment-facade fire. A compartment model of 1.2 m × 0.9 m × 0.9 m is established in this study, and an air curtain is facilized at the compartment opening for fire protection. The air curtain jet velocity, the heat release rate as well as the opening dimension are changed as representative to typical fire scenarios. Flame behavior and heat flux profile of spill plume are captured by CCD cameras and water-cooled heat flux gauge for discussion. Results show that the air curtain jet could have a substantial impact on the flame behavior of spilled plumes from a compartment-facade fire. Compared to that of free condition without air curtain, the flame height becomes 20 % lower as the jet velocity to be 2 m/s, while it becomes 60 % lower as the jet velocity to be 5 m/s. At the meantime, the spilled plume is noticeably influenced by the “restricting effect” of the air curtain and thus flame trajectory is significantly flattened by the inertia force of the air curtain but also pushed farther from the facade wall, resulting in an increasing of flame depth. The decreasing in flame height plus the increasing flame depth would result in the reduction of heat flux intensity upon the facade wall, which is helpful for the fire protection of the facade wall. Then, a normalized factor of Υ is further proposed to the new correlation of flame height under various air curtain jet conditions at the opening. Regarding to different ejected flame behavior at the presence of air curtain jet, the heat flux upon the facade wall is quantitative analyzed to evaluate the thermal impact of spilled plume from the compartment fire. Finally, new correlations of the vertical heat flux upon the facade wall could be well achieved by a function of normalized height factor Z/Zf, v.

Facade flame depth coming out from the fire compartment opening subject the external sideward wind

When a fuel leakage/energy release (i.e., combustion, fire) inside a compartment, the external facade flame characteristics through opening are of great significance for protecting human life and property, as well as assessment of its risk and impact to urban environment. This work investigates facade flame depth coming out from fire compartment opening with fuel combustion/energy release inside the compartment for under-ventilated condition under sideward wind, which is a fundamental parameter in describing facade morphological characteristics in particular building fire, meanwhile still no work can be found in previous studies. Results show that, as sideward wind speed increases, the flame depth first increases a bit then decreases significantly. A CFD simulation is carried out to understand flow field controlling facade flame depth under windless and sideward wind situations. A physical analysis is presented on account of mechanisms of complex interaction among sideward wind inertial, outflow momentum through opening, and flame buoyancy flux are proposed based on controlling mechanisms, which well represents the facade flame depth by a non-dimensional function under sideward wind condition. This study makes an important new contribution comparing with previous knowledge, and has potential practical value for fuel combustion inside confined space and the corresponding numerical model verification.

The 1986 Annaburroo experimental grassland fires: data

Background In 1986, CSIRO conducted a large program of experimental fires in grassland at Annaburroo Station, Northern Territory, Australia, with the objective of quantifying the effect of fuel condition (load and height) on fire behaviour. Aims This paper provides the data collected during this program, representing a unique set of observations and measurements of large, free-burning experimental fires conducted in a multi-factor experimental design. Methods Data are collated by experimental burn plot, providing detailed measurements of weather (wind speed, air temperature, relative humidity), fuel state (load, height, moisture content, curing) and fire behaviour (rate of spread, flame depth, flame height, head fire width), as well as processed information (e.g. steady-state rate of spread). Data availability The data are made available for free download on the CSIRO Data Access Portal (https://data.csiro.au/collection/csiro:58746) and include detailed metadata descriptions of the data and their structure, also provided in this article. Conclusions We have made the data available for fire behaviour researchers around the world to use in their research under the Creative Commons Attributions licence. It is hoped they will analyse these data and extract new and innovative insights to help improve our understanding of wildland fires burning in grass fuels.

Experimental investigation on morphology of window-ejected flame without facade effect

As the key characteristic of window-ejected flames, flame morphology such as flame height, flame depth involves the ignition of surrounding buildings. This article experimentally investigated the flame height, the maximum flame width and the maximum flame depth ejected from a fire compartment without the facade wall effect. A small-scale compartment with a sidewall opening attached to compartment top is used for simulating the situation without the facade wall. The ejected flame height, maximum flame width and maximum flame depth were recorded and measured for various opening sizes and fuel supply rates. Results show that: (1) The ejected flame height increases as the increase of fuel supply rate and decreases as the increase of opening dimension, and was obviously lower than that with the facade effect; (2) The maximum flame width increases as the increase of fuel supply rate and decreases with the increase of opening dimension, and was higher than that with the facade effect; (3) The maximum flame depth increases as the increase of fuel supply rate, and the formation of maximum flame depth was analyzed based on the ejected thermal plume momentum and the air entrainment momentum. A dimensionless parameter, which presents the complex competition mechanism between the plume momentum and the air entrainment momentum, was proposed to characterize their impacts on flame depth, by which the maximum flame depth was represented well. This work would provide theoretical support for building fire prevention and control.

Experimental study of ejected fire plume from an opening under facing wind passing through the roof

This paper presents an experimental study and analysis of the ejected fire plume from an opening at the top floor of the compartment under facing wind condition. The experiments are conducted by a reduced-scale cubic compartment. An opening facing the wind tunnel exit is set at the center of the compartment side wall. The temperature at the opening and inside the compartment, critical heat release rate for flame ejection, ejected flame morphologic characteristic parameters are measured and quantified by thermocouples and a digital video camera. It's found that the temperature and critical heat release rate for flame ejection first increase then decrease with increasing of facing wind speed. The flame height, flame depth and flame horizontal length increase with heat release rate. The flame depth decreases, flame horizontal length increases as the facing wind speed, while flame height shows a complex evolution with facing wind. A physical model is developed considering mechanisms of flame ejection initial horizontal momentum, flame buoyancy, wind momentum and non-dimensional mass fuel supply rate. The proposed model correlates the experimental data of flame morphologic parameters well. This work considering the practical fire scenario provides an important knowledge supplement on the classical building fire dynamics.

A reduced-scale experimental investigation of facade flame behavior ejected from a top-hung window of fire compartment

In case of a room fire, facade flame ejected from a window could lead to catastrophic consequence. Top-hung window is a common feature in modern buildings. This paper investigates, for the first time, the ejected facade flame behavior from a top-hung window. Experiments are performed by a cubic scale model compartment with a top-hung window of four opening angles and four dimensions under various heat release rates (HRR). A total of 119 test conditions are involved to quantify generally the ejected facade flame behavior and morphological characteristics, characterized by its vertical height and horizontal depth (normal to facade). It is found that the position of neutral plane between the inflow and outflow at the top-hung window increases with opening angle, while being smaller as the window width is larger or the window height is smaller. The critical HRR for flame ejection is higher as the top-hung window opening angle is larger. The facade flame depth increases, while the flame height decreases, with increasing of opening angle of the top-hung window. A new characteristic length taking into account both the top-hung window dimensions and window opening angle is deduced by the mass balance analysis of the inflow and outflow as well as ventilation factor through the top-hung window. New dimensionless models are established to describe the critical HRR for flame ejection, the height of the facade flame and flame depth as a function of the newly derived top-hung window ventilation factor as well as characteristic length, showing good fitting of experimental results. The present study provides the basic data, understanding and model of facade flame characteristics from a top-hung window of a fire compartment, which is essential in estimation of its risk and adverse impact to urban environment as a new supplementary over previous knowledge limited for freely opened windows.

An experimental study on mechanism of eruptive fire

ABSTRACT This paper presents a systematical experimental study on eruptive fire over a pine needle fuel bed in trench configuration in still air to elucidate the essential mechanism of eruptive fire. The fire spread characteristics under different slope angles (0–40°) and trench aspect ratios (0.1–0.4) are investigated by a comprehensive analysis of experimental data. The data of fire spread rate and flame depth reveal that eruptive fire is a kind of “sudden accident” that happens when the slope angle reaches a certain critical value. Under such a circumstance, as the flame depth increases after ignition, the fireline intensity increases and finally triggers the fire eruption when it comes to a necessary level. The fire eruption induces a series of changes in burning and fire spread behaviors such as sharp increases in velocity and influence range of the attached flow in the direction of fire spread. Due to the hot gas flow induced by flame attachment, compared to flame radiation, convective heating becomes to play a comparative or even more critical role in fire spread. The influence range of the attached flow increases with the slope angle or aspect ratio of the trench configuration, while for a given terrain slope, there is a ceiling effect of the trench aspect ratio on the fire eruption behavior.

An experimental investigation into the wildland fire burning characteristics of loblolly pine needles

Wildland fires present a threat to both the environment and to homes and businesses in the wildland urban interface. Understanding the behavior of wildland fires is crucial for developing informed risk management techniques, such as prescribed burning, to prevent uncontrolled fires, which devastate communities globally. In this work, an optically accessible facility was developed combining traditional and optical diagnostic techniques to study the combustion of wildland fuels. This facility was then used to study the burning characteristics of loblolly pine needles (Pinus taeda). Mass loss rate, flame temperature, propagation rate, flame geometry, and OH* chemiluminescent flame intensity were measured for fuel loadings of 0.98, 1.31, and 1.63 kg/m2. Mass loss rate increased linearly with fuel loading. The flame propagation rate increased slightly with fuel loading while approaching an asymptotic value. The flame height and surface area increased more significantly with fuel loading, but also approached asymptotic values. Flame depth increased slightly with fuel loading, and throughout the duration of the test. The flame intensity and flame temperature did not change significantly with fuel loading.

Secondary air distribution in a 600 MWe multi-injection multi-staging down-fired boiler: A comprehensive study

Multi-injection multi-staging combustion technology was designed to provide high boiler efficiency together with low pollutant emissions when employing hard-to-burn coal, and showed good application prospects. In this technology, to achieve the first staged air and more flexible combustion regulation on the arches, the secondary air employs a more complicated distribution surrounding the coal/air flows. This design concept significantly affects the flow and combustion characteristics of the actual boiler. Elucidating the underlying mechanisms by which the various secondary air parameters affect the boiler operation requires a comprehensive analysis of the secondary air distribution on the arches. In this work, small-scale cold modeling tests combined with industrial-scale trials using two actual 600 MWe down-fired boilers were conducted. The results show that the secondary air distribution can significantly affect the near-burner flow and mixing as well as the decay of the downward airflow velocity. High velocity secondary air, especially the secondary air surrounding the fuel-rich flow, evidently promotes the advanced mixing of fuel-rich and fuel-lean flows. Increasing the secondary air flux around the fuel-rich flow and raising the secondary air velocity decrease the pulverized coal heating rate, while a decreased inner secondary air velocity advances the coal ignition. Adjusting the secondary air also greatly affects the downward flame depth and the unburnt coal proportion in the near wall region of the furnace hopper, such that the flame kernel location and the temperatures near the hopper wall are modified. In addition, NOx formation can be accelerated by a high secondary air flux near the fuel-rich flow or a large contact area between the secondary air and the fuel-rich flow.

Factors affecting the downward flame depth in a 600 MW down-fired boiler incorporating multiple-injection and multiple-staging technology

Considering the excessively deep flame depth existing in a 600 MW down-fired boiler incorporating multiple-injection and multiple-staging technology, 1:20 scale aerodynamic tests were conducted to incrementally improve factors responsible for the deep flame depth. These trials demonstrated that, in all cases, the downward velocity near the wing walls decayed more rapidly than that near the furnace center. Increasing the mass ratio of pulverized coal in fuel-rich flow to that in fuel-lean flow and reducing the secondary air ratio while simultaneously increasing the tertiary air ratio was found to increase the penetration depth. Increasing the distance between adjacent burners, optimally lowering the fuel-rich and fuel-lean flow velocities, and reducing the secondary air velocity all decreased the penetration depth. The comprehensively improved downward airflow depth and air flux into the furnace hopper were reduced by 6.3% and 14.9%, respectively. Cold-state airflow tracing tests were performed in an actual boiler, the improved downward airflow depth reduced apparently, which was consistent with the results from modeling tests. Industrial-scale hot-state experiments determined improved hopper near-wall temperatures of 700–800 °C near the furnace center (values that were lower than those of 800–900 °C near the wing walls). These values were approximately 450 °C less than the prior temperatures at the same location, indicating the flame penetration depth was greatly reduced.

Upslope fire spread over a pine needle fuel bed in a trench associated with eruptive fire

Eruptive fire is a typical extreme fire behavior in wildland fuels, characterized by a sudden acceleration of fire spread over confined slopes in trench-like terrain (e.g. in a deep canyon). However, wildfire spread acceleration in trench configuration has received little study. This paper presents a systematic experimental study on the fire spread over a pine needle fuel bed in an inclined trench. The effects of aspect ratio (ratio of the trench side wall height to the trench width) and slope angle on the fire spread behaviors are investigated. It is indicated that flame attachment toward the fuel bed surface is induced by the combined effects of the slope and the air entrainment restriction due to the trench configuration. The flame attachment occurring at higher slopes with higher aspect ratios induces an acceleration of fire spread, causing the fire spread to undergo a rapid transition from a steady phase with lower rate of spread (ROS) to quick spread phase with much higher ROS. The fire spread acceleration induced by the flame attachment leads to remarkable enhancement of burning intensity, as verified by long flame depth and high mass loss rate during fire spread. The fuel consumption efficiency decreases almost linearly with increasing slope angle. Under higher slope angles (higher than 20○ in this work) the flame tilting induces hot gas flow ahead of the flame front, which enhances the convective heating. When flame attachment occurs, the spatial influence range of convective heating increases sharply. It is concluded that for eruptive fire, radiative heating and convective heating play a comparable role in fuel preheating. Experimental data suggest that when flame attachment occurs the convective heating is greatly enhanced and becomes an important mechanism of fuel preheating. This is inferred to be a major potential mechanism for eruptive fire in trench configuration.

Experimental study on vertical fire spread along a wooden façade

The main purpose of these tests is to get some information on fire spread along a combustible facade in order to assess the ability of numerical codes [1] to simulate flame propagation. Many authors have already studied compartment fire with external flames for non-combustible facades [2,3,4] and also combustible facades [5,6]. Studies on non-combustible facade allow to describe flames (flame depth, flame height, thermal actions, etc.) and to explain the influence of geometrical parameters (opening dimensions, U configuration, etc.) on flame shape and flame behaviour. These studies highlight the importance of some of these parameters on vertical spread and correlations are issued from these experimental studies. If many researches on combustible facades exist, it is difficult to find academic configurations with a large instrumentation and generally, these studies focus on a facade system involving complex combustible materials (insulation or composite). That's why it was decided to perform some large scale tests to study fire spread along a vertical wood wall. © Owned by the authors, published by EDP Sciences, 2016 Language: en

Window ejected flame width and depth evolution along facade from under-ventilated enclosure fires

This paper investigates window ejected flame width and depth evolutions along facade from under-ventilated enclosure fires. Experiments are carried out by 1:4 scale model. Two CCD cameras are employed to record the evolutions of flame depth and width. The flame base position (vertical height above the bottom of the opening), flame depth (width) along with their maximum values and corresponding positions (vertical height above the flame base position) are measured and analyzed by non-dimensional scaling. It is found that the flame base position is independent of fire heat release rate and its ratio to opening height is nearly the same. The flame depth for all heat release rates and the flame width for openings with aspect ratio in the range of 0.5≤W/H≤1.5 (the “(semi-) axi-symmetrical flame type”) first increases, reaching a maximum value and then decreases with height. However, for the opening with a relative larger aspect ratio (W/H=2) (“wall flame type”), the flame width decreases monotonously with height. The maximum flame width and its corresponding vertical position for “(semi-) axi-symmetrical flame type” are found to be well correlated by characteristic length scale ℓ1 [=(AH)2/5] non-dimensionally after normalized by the flame height. Meanwhile, the maximum flame depth for all conditions is found to be well correlated by characteristic length scale ℓ2 [=(AH2)1/4] non-dimensionally after normalized by the flame height.

Experimental confirmation of the MWIR and LWIR grey body assumption for vegetation fire flame emissivity

The temperature and emissivity of forest fire flames play a key role in understanding fire behaviour, modelling fire spread and calculating fire parameters by means of active fire thermal remote sensing. Essential to many of these is the often-made assumption that vegetation fire flames behave as grey bodies in the infrared (IR). Although the emissivity of flames and its relationship to flame depth has been measured experimentally using thermal imagers working in the long-wave IR (LWIR, 8–12µm), no published study has yet demonstrated relationships in the important mid-wave IR (MWIR, 3–5µm) spectral region, nor conclusively demonstrated that assumptions about grey body behaviour across these two important IR atmospheric windows fit well with reality. Our study explores these issues using measurements of boreal forest fuels burned with flame depths ranging from 0.2 to 4.2 m. Observations of two stable black body sources made through the differing flame depths were used to explore flame spectral emissivities and their relationship to flame depth. We found essentially the same relationship between flame emissivity and flame depth for both spectral regions, (extinction coefficient K=0.7 m–1), confirming that the grey body assumption for forest fire flames in the MWIR and LWIR atmospheric windows appears valid for the fire conditions encountered here.

Effect of Sample Width on Characteristics of Flame Spread across Eucalyptus Wood Surface

To study the effect of sample width on flame spread characteristics, a series of laboratory-scale experiments were conducted employing eucalyptus wood with width from 3 to 7 cm. Flame dimension, flame spread rate and mass loss rate were obtained. The relationship between these flame spread characteristics and sample width was explored. Both the dimensionless average flame height and depth vary as the-n power of sample width. With the increase of sample width, both the flame spread rate and mass loss rate first decrease and then rise. The minimum values appear when sample width measures 6 cm.

Metrological Tool for the Characterization of Flame Fronts Based on the Coupling of a Heat Flux Approach with Image Processing Data

The aim of this paper is to present some improvements in the metrology of forest fire flames by coupling image processing and radiative heat flux measurement. A new metrological tool using a visual video camera and a specific multiple thermal sensors is proposed. By means of an appropriate segmentation algorithm and the Direct Linear Transformation, the image processing methodology gives the forward or the backward fire front positions as input data to a radiative heat flux approach. Using a simplified flame model, this technique provides fire front positions versus time, and average values for the flame length, flame depth, flame tilt angle, apparent flame temperature and flame emissivity. The rates of spread, obtained by a linear regression of the determined fire front positions, as well as the thermal and radiative properties of the flame, are compared favorably to those given in the literature.

Numerical study of wind effects on the characteristics of flames from non-propagating vegetation fires

A variant of a three-dimensional two-phase model was proposed to address the central question of how wind affects the geometric properties of flames from steady vegetation fires. The pyrolysis rate and the spatial distribution of burning vegetation were representative of the heat release rates of fires ranging from single tree or tree group “torching” to continuous surface fires. Wind speed, U ∞ , was varied from 0.5 to 8 m s −1. Results revealed that about one third of the heat produced by combustion is released within the fuel bed, and this occurs almost independently of wind speed and vegetation distribution. Predicted values of flame height, flame length and tilt angle were found to be consistent with available literature data. Model results showed that the flame length and height normalized by the flame depth, D, scale with the modified Froude number Fr D = U ∞ 2 / g D and the dimensionless heat release rate. Predicted mean velocity flow streamlines appeared to be nearly straight in both the flaming region and the inert plume, supporting the relevance of a far field analysis to determine the flame tilt angle. This analysis showed that the tangent of the flame tilt angle may be expressed in terms of the Froude number based on the characteristic length scale of the plume. The numerical model was then used to quantify these relationships. The relevance of scaling parameters was supported by both field and laboratory-scale experiments.

Understorey fire propagation and tree mortality on adjacent areas to an Amazonian deforestation fire

Fire characteristics in tropical ecosystems are poorly documented quantitatively in the literature. This paper describes an understorey fire propagating across the edges of a biomass burn of a cleared primary forest. The experiment was carried out in 2001 in the Amazon forest near Alta Floresta, state of Mato Grosso, Brazil, as part of biomass burning experiments conducted in the same area since 1997. The vegetation of a 200 × 200-m2 forested area was clear-cut in early June and burned in late August. The understorey fire that escaped from the main burn was monitored across the four sides of the land clearing area. Flame-front spread varied between 0.14 and 0.35 m min–1. Maximum flame height was about 30 cm and typical flame depth was 10 to 15 cm. Tree mortality was investigated in 2003 in four areas adjacent to the biomass burning experiment. A total of 210 trees were counted in the four areas, 29.5% were dead as a consequence of the understorey fire that had occurred 2 years before. This fire-caused mortality is evidence of the synergistic effect between slash burning, tree mortality and future fire vulnerability on the forest–land clearing interfaces.

Fire eruption through intensity and spread rate interaction mediated by flow attachment

Conditions under which a fire has a stable steady spread rate or under which it is able to spread eruptively up a slope or in confined topography are of considerable interest from a practical and safety point of view. The physical interactions that give rise to either form of behaviour are investigated by way of a mathematical model, in which different expressions for the rate of feedback from intensity into spread-rate are found to identify a threshold between eruptive and stable spread of a line fire. In turn, changes in the fireline intensity in any unsteady evolution are mainly determined by the history of the spread-rate over a burnout time (in effect, by changes in flame depth). Under stable conditions, any initial spread-rate evolves towards the steady spread rate on a time-scale of the order of the burnout time. But above the threshold, eruptive fire-growth sets in; the spread-rate and intensity both grow indefinitely. It is argued that a change in the nature of the flow field around a line fire engenders the change from one form of behaviour to another. If the air flow separates at the flame, so that air is drawn away from the ground and vegetation surface, then the model provides strong reasons to expect that the steady spread rate is stable. On the other hand, laboratory experiments and a controlled field burn confirm that eruptive behaviour is more likely to be associated with flow attachment. As a result, if the air immediately ahead of a fire that is spreading uphill, flows up the slope away from the fire then conditions arise for a potentially very dangerous acceleration in the spread-rate of the fire, along with a corresponding growth in its intensity. As is shown by the experiments, this form of air-flow can be generated by the fire itself without any change in external conditions such as ambient wind.

Dissociation energies and excitation levels of alkaline-earth oxides

AMBIGUITY about the dissociation energy of alkaline-earth oxides, as derived from flame photometric measurements, originates from the uncertainty with regard to the emitter of the band spectra, the designation and energy of the ground and excited state levels, respectively, the a’-parameter of the atomic resonance lines involved, and the possible side-reactions of these elements in the flame. Possible side-reactions may lead to the formation of hydroxides, which may also give visible band emissions. We have tried to eliminate the uncertainty about the emitter by using special “dry” CO-air flames (containing virtually no hydrogen (-compounds)), comparing the results with those of similar (“moist”) flames into which a small amount of water vapour was introduced. Alkalineearth salt was fed into the “dry” flame by dry evaporation of the salt in a heated pot placed in the Na conduit to the flame. Besides, the height of the upper excitation levels of the bands observed could be deduced from the measured temperature-dependence of the band intensities, while correcting for the (slight) dependence on temperature of the metal oxide concentration and flame depth. Plotting the observed intensities semilogarithmically against reciprocal temperature directly yields the excitation energy. The excitation energies found are listed in Table 1 and compared with data from the literature. It appears that the lower state of the transitions involved is by O-6 to 1.0 eV higher than the ground state for all alkaline-earth oxides investigated (cf. last column in Table 1).